Prostate cancer is tlie most frequently diagnosed non-cutaneous cancer and is the second leading cause of cancer death in American men. Stratifying patients with high-risl< of prostate cancer recurrence following primary therapy from those of low-risk disease at an early stage may support curative adjuvant-therapy for prostate cancers of metastatic potential. In a nested case control study, CX3CL1, IL-15, and CCL4 expression were independent predictors of recurrence status. This proposal tests the hypothesis that chemokine biomarkers that predict biochemical recurrence of prostate cancer regulate metastatic progression of the cancer and curcumin can inhibit metastasis of prostate cancer. In Aim 1 CX3CL1, CCL4 and IL-15 will be expressed in prostate cancer epithelial cell lines to determine whether cancer cells expressing recombinant chemokines CX3CL1 and IL-15 (biochemical recurrent-free patients) and CCL4 (biochemical recurrent patients) can differentially regulate metastasis of prostate cancer in vitro. Xenograft tumor models will test whether tumors expressing recombinant chemokines CX3CL1, IL-15 and CCL4 can differentially regulate metastasis of prostate cancer. Aim 2 will identify the role of inflammation on prostate cancer progression. Tgfbr2'*'^ prostates that develop into adenocarcinoma will be grafted with stromal cells that express CX3CL1, CCL4 and IL-15. Bone marrow derived cells that are recruited to the prostate will be identified and quantified. Chimeric mice with bone marrow derived cells knocked out for the expression of the primary CX3CL1 receptor and CCL4 receptor will be developed. The goal will be to characterize the consequences of ablating the recruitment of specific populations of inflammatory cells to the prostate. In Aim 3 prostate cancer cell lines expressing CX3CL1, CCL4 and IL-15 will be treated with curcumin to determine whether curcumin can inhibit metastasis of prostate cancer in vitro through the regulation of CX3CL1, CCL4 and IL-15. Xenograft tumor models will test whether curcumin modulates tumor metastasis in vivo through the regulation of CX3CL1, CCL4 and IL-15. These results will provide biologic basis for the clinical use of the chemokine biomarkers and development of new therapies aimed to inhibit prostate cancer metastasis. RELEVANCE (See instructions): The early determination of metastatic prostate cancer can enable appropriate treatment and save patients who would eventually die of recurrence of prostate cancer. The proposed research will determine the function of predictive chemokines and curcumin in prostate cancer metastasis. This might allow for the future development of new therapies aimed to inhibit the growth and spread of prostate cancer tumors.